Get to know your Profs with Manolis Savva
Manolis Savva is an assistant professor in the School of Computing Science and a Canada Research Chair in Computer Graphics. Before his current position, he was a visiting researcher at Facebook AI Research and a postdoctoral researcher at Princeton University. He received his Ph.D. from Stanford University under the supervision of Pat Hanrahan and his B.A. in Physics and Computer Science from Cornell University.
Professor Savva’s research focuses on analyzing, organizing, and generating 3D content. The methods he works on are stepping stones towards holistic 3D scene understanding revolving around people, with applications in computer graphics, computer vision, and robotics. His works have been recognized through several awards, including the best paper nomination (Habitat, ICCV 2019), two SGP dataset awards (ShapeNet, SGP 2018; ScanNet, SGP 2020), and the 2022 GI Early Career Researcher Award.
We had a coffee chat with Manolis to know more about him.
What influenced your decision to study computer science, and why did you focus on visual computing?
In my undergraduate days, I studied physics and got involved in research at a low-temperature physics lab. The types of experiments we would do would take several months to prepare complex experimental equipment for near-zero temperature measurements of how materials behave. As an undergrad researcher, I developed and fabricated a “resonator cavity” device to help measure electrical signals at these low temperatures. Developing the "resonator cavity" involved using 3D CAD software to design, simulate, and fabricate the device. This work influenced my interest in the discipline of 3D computer graphics, which studies the algorithms involved in 3D design, rendering and simulation. The area was a fascinating combination of my interests in physics, math, and computation, so I decided to focus on it for the remainder of my undergraduate degree.
What’s your favorite thing about teaching and doing research at SFU?
The School of Computing Science at SFU is a very vibrant department. It gives me an overall sense of a “young” institution, primarily because there are great many new faculty members, new degree programs, and new student cohorts in the past few years. This is an energizing atmosphere both from the teaching and research perspectives.
What is your most recent research focus or project you are working on?
Most recently, I have been working on various projects to acquire high-fidelity 3D representations of actual rooms and how people can interact with objects in those rooms. 3D reconstruction of dynamic objects like kitchen cabinets while being manipulated by a person is still a challenging research question. Better algorithms for achieving such dynamic reconstruction have many applications for virtual and augmented reality systems and for democratizing 3D content creation.
What has been your proudest scholarly achievement so far?
It was an honor to receive an award as the Canada Research Chair in 2020 and the GI Early Career Researcher Award in 2022. I never expected such recognitions so early in my career as an academic.
What is your favourite programming language and why?
I honestly can’t pick a specific favourite. Each language has its tradeoffs and is a good choice for different domains and types of problems. Nowadays, Python is the dominant language for ML and AI researchers. However, implementation of performance-critical code is best done in C and C++ or more modern compiled languages such as Rust. Web-based codebases require the use of JavaScript, which is quirky but also quite robust and increasingly performant.
What will you say fueled your success in graduate school?
If there is one thing to which I would attribute my success as a grad student, it is picking up the habit of task and time planning for research projects. Often, there is a misconception that research is all about random “inspiration.” In most ways, I believe it is the opposite: steadiness, planning, and regularity in working habits win out in the longer term.
What will be your advice to current and future graduate students?
Graduate school can be isolating. I will tell students to learn the value of talking with as many people as possible at all levels (peers, juniors, and mentors). We can all learn something from others no matter their background and experience level.
Do you have any fond memories of your time at SFU?
One of my fondest memories at SFU is teaching a newly developed “Frontiers of Visual Computing” class for the Visual Computing program in my first year as a faculty member.